1

Supplementary Information: Meta-analysis of exome array data identifies six novel genetic loci for lung function

Victoria E Jackson, Jeanne C Lautourelle, Louise V Wain, Albert V. Smith, Megan L. Grove, Traci M Bartz, Maen Obeidat, Michael A Province, Wei Gao, Beenish Qaiser, David J Porteous, Patricia A Cassano, Tarunveer S.

Ahluwalia, Niels Grarup, Jin Li, Elisabeth Altmaier, Jonathan Marten, Sarah E Harris, Ani Manichaikul, Tess D Pottinger, Ruifang Li-Gao, Allan Lind-Thomsen, Anubha Mahajan, Lies Lahousse, Medea Imboden, Alexander

Teumer, Bram Prins, Leo-Pekka Lyytikäinen, Gudny Eiriksdottir, Nora Franceschini, Colleen M Sitlani, Jennifer A Brody, Yohan Bossé, Wim Timens, Aldi Kraja, Anu Loukola, Wenbo Tang, Yongmei Liu, Jette Bork-Jensen,

Johanne M. Justesen, Allan Linneberg, Leslie A. Lange, Rajesh Rawal, Stefan Karrasch, Jennifer E Huffman, Blair H Smith, Gail Davies, Kristin M Burkart, Josyf C Mychaleckyj , Tobias N. Bonten, Stefan Enroth, Lars Lind, Guy G

Brusselle, Ashish Kumar, Beate Stubbe, Understanding Society Scientific Group , Mika Kähönen, Annah B. Wyss, Bruce M Psaty, Susan R Heckbert, Ke Hao, Taina Rantanen, Stephen B Kritchevsky, Kurt Lohman, Tea Skaaby, Charlotta Pisinger, Torben Hansen, Holger Schulz, Ozren Polasek, Archie Campbell, John M Starr,

Stephen S Rich, Dennis O. Mook-Kanamori, Åsa Johansson, Erik Ingelsson, André G Uitterlinden, Stefan Weiss, Olli T. Raitakari, Vilmundur Gudnason, Kari E. North, Sina A Gharib, Don D Sin, Kent D Taylor, George T

O’Connor, Jaakko Kaprio, Tamara B Harris, Oluf Pedersen, Henrik Vestergaard, James G. Wilson, Konstantin Strauch, Caroline Hayward, Shona Kerr, Ian J Deary, R Graham Barr, Renée de Mutsert, Ulf Gyllensten, Andrew

P Morris, M. Arfan Ikram, Nicole M Probst-Hensch, Sven Gläse, Eleftheria Zeggini, Terho Lehtimäki, David P Strachan, Josee Dupuis, Alanna C. Morrison, Ian P Hall, Martin D Tobin, Stephanie J London

Supplementary Note

Individual study descriptions This section describes study-specific characteristics, and details of spirometric and other measurements. All

participants provided written informed consent and studies were approved by local Research Ethics Committees and/or Institutional Review boards.

The Reykjavik Study cohort originally comprised a random sample of 30,795 men and women born in 1907–

1935 and living in Reykjavik in 1967. A total of 19381 attended, resulting in 71% recruitment rate. The study sample was divided into six groups by birth year and birth date within month. One group was designated for

longitudinal follow-up and was examined in all stages. One group was designated a control group and was not included in examinations until 1991. Other groups were invited to participate in specific stages of the study.

Between 2002 and 2006, the AGES-Reykjavik study (AGES) re-examined 5764 survivors of the original cohort who had participated before in the Reykjavik Study.

Atherosclerosis Risk in Communities (ARIC), is a population based study of risk factors for atherosclerosis and

its sequelae1 in adults from four U.S. field centers aged 45-64 at recruitment in 1987-1989. ARIC spirometry measurements were made with a Collins Survey II water-seal spirometer (Collins Medical, Inc.) and Pulmo-

Screen II software (PDS Healthcare Products, Inc.).

Details of the British 1958 Birth Cohort (1958BC) biomedical follow-up have been previously reported2.

Spirometry at age 44–45 years was performed in the standing position without nose clips, using a Vitalograph handheld spirometer as previously described3. In the analysis, all readings with a best-test variation greater

than 10% were excluded.

The Cardiovascular Health Study (CHS) is a population-based cohort study of risk factors for coronary heart disease and stroke in adults ≥65 years conducted across four field centers.4 The original predominantly

European ancestry cohort of 5,201 persons was recruited in 1989-1990 from random samples of the Medicare

2

eligibility lists; subsequently, an additional predominantly African-American cohort of 687 persons was enrolled in 1992-1993 for a total sample of 5,888. The baseline exam consisted of a clinic examination that

included assessment of height and weight and self-reported smoking information. Pulmonary function testing was conducted by trained spirometry technicians. FEV1/FVC and FEV1 measures met American Thoracic Society

criteria for acceptability. The pulmonary function measures analyzed were from the baseline visit for the original cohort and from one year after baseline for the second cohort. Measurements were made with a

Collins Survey I water-seal spirometer (Collins Medical, Inc.) and software from S&M Instruments.5 Blood samples were drawn from all participants at their baseline examination and DNA was subsequently extracted

from available samples. Genotyping was performed at the General Clinical Research Center’s Phenotyping/Genotyping Laboratory at Cedars-Sinai among CHS participants who consented to genetic testing

and had DNA available using the Illumina HumanExome BeadChip v1.0 (the "Exome Chip") using standard protocols. Cryptically related individuals were identified and only one individual from each related group was

retained for analysis.

The CROATIA study was initiated to investigate the use of isolated rather than urban populations for the

identification of genes associated with medically-relevant quantitative traits. Three cohorts have been recruited as part of the CROATIA study, of which one, CROATIA-Korcula 6 has been used in these analyses.

CROATIA-Korcula was recruited from 2007 to 2008 from the town of Korcula and the villages of Lumbarda, Zrnovo and Racisce on the island of Korcula, Croatia with 969 adults aged 18-98 agreeing to participate.

Participants donated blood for DNA extraction and biochemical measurements as well as undergoing some anthropometric measurements and physiological tests to measure traits such as height, weight and blood

pressure, and finally completing several questionnaires relating to general health, medical history, diet and lifestyle. Ethical approval was obtained from appropriate regulatory bodies in both Scotland and Croatia and participants gave informed consent prior to joining the study.

Details of the design of the NHLBI Family Heart Study (FAMHS) have been described previously. 7Spirometry was performed during the participant's clinical examination using a computerized volume-based spirometer

and was reported at body temperature and pressure, saturated. FEV1 and FVC were measured.

The Framingham Heart Study (FHS) is a longitudinal community-based family study that originated in 1948 with the recruitment of adults from the town of Framingham, MA. The offspring of the original cohort were

recruited to participate in 1971, and the third generation (Gen3) was recruited starting in 2002. Spirometry has been measured on all three generations of the participating families as part of the clinical examinations. For

the original and offspring cohorts, spirometry was performed using a 6-L water-filled Collins survey spirometer connected to an Eagle II microprocessor (Collins Medical, Braintree, MA) or in later offspring examinations to a

personal computer running software developed by S&M Instruments, Doylestown, PA. For the Gen3 cohort, spirometry was performed using the CPL System 53 (Collins Medical, Braintree, MA) with a dry rolling-seal

spirometer. All of these systems provided an automatic correction for body temperature and pressure, saturated, and provided quality assurance information.

The Finnish Twin Cohort (FTC) sample originates from The Finnish Twin Study of Ageing (FITSA) sub-study.

Participants were recruited from the older Finnish Twin Cohort for a clinical study of functional limitations in older women. Clinical assessment was conducted in 2000-2001 at the University of Jyväskylä. The final sample

consisted of 103 monozygotic (MZ) and 114 dizygotic (DZ) twin pairs. Lung function was measured in the standing position using an electronic spirometer (Medikro 202, Kuopio, Finland). The subject was asked to

inhale maximally and to exhale as fast as possible into a mouthpiece connected to a flow transducer and a flow-volume curve was created. At least two tests were performed and the best result taken for the analyses.

Spirometer was calibrated daily with a three-litre pump and was accurate to within 1%.

The Generation Scotland: Scottish Family Health Study (GS:SFHS)is a collaboration between the Scottish Universities and the NHS, funded by the Chief Scientist Office of the Scottish Government. GS:SFHS is a family-

3

based genetic epidemiology cohort with DNA, other biological samples (serum, urine and cryopreserved whole blood) and socio-demographic and clinical data from ~24,000 volunteers, aged 18-98 years, in ~7,000 family

groups. Participants were recruited across Scotland, with some family members from further afield, from 2006-2011. Most (87%) participants were born in Scotland and 96% in the UK or Ireland. The cohort profile has

been published8. GS:SFHS operates under appropriate ethical approvals, and all participants gave written informed consent. Generation Scotland is a collaboration between the University Medical Schools and National

Health Service in Aberdeen, Dundee, Edinburgh and Glasgow (UK).

The Health Aging and Body Composition (HABC) study is a prospective observational cohort of well-

functioning individuals aged 70–79 years, which recruited 3,075 community-dwelling African and European Americans, men and women, at two field centers at the University of Pittsburgh, Pennsylvania and the

University of Tennessee, Memphis. Spirometry was performed with a horizontal dry rolling seal spirometer (SensorMedics Corporation, Yorba Linda, CA) connected to a computer. Pulmonary function testing followed

ATS guidelines for the standardization of spirometry, and is described in detail elsewhere.9 Health ABC genotyped 1,794 self-described white participants and 1,281 self-described black participants at baseline with

available DNA and consent to genetics testing; of these 1,661 white participants and 1,139 black participants passed quality control benchmarks (call rate > 97%, no sex mismatch, and cryptic relatedness), and 1,457 and

943, respectively, had pulmonary function measurements and complete data on covariates.

Health2006 is a population-based epidemiological study of general health, diabetes and cardiovascular disease of individuals aged 18-74 years conducted at the Research Centre for Prevention and Health in Glostrup,

Denmark.10 Spirometry was performed with the Spiro USB Spirometer (MicroMedical Limited, Rochester, Kent, UK). Measurements of forced expiratory volume in one second (FEV1) and forced vital capacity (FVC) were

recorded.

Health2008 is an extension of the Health2006 study, where a random sample of the general population aged 30 to 60 years from the same regional areas in Copenhagen County was drawn from the Civil Registration

System. Methodology was the same as in Health2006.

The Inter99 study carried out in 1999-2001 included invitation of 12,934 persons aged 30-60 years drawn from an age- and sex-stratified random sample of the population.11 The baseline participation rate was 52.5%, and

the study included 6784 persons. The Inter99 study was a population-based randomized controlled trial (CT00289237, ClinicalTrials.gov) and investigated the effects of lifestyle intervention on CVD. Spirometry was

performed with the Cardiosoft® software (GE Medical Systems, Freiburg, Germany) and an LF501 respiration flow transducer (Erich Jaeger B.V. and Marquette Hellige GmbH, Freiburg, Germany). Each morning, the

spirometers were checked with a 3-L syringe. Measurements of forced expiratory volume in one second (FEV1) and forced vital capacity (FVC) were recorded. Spirometry was performed by trained healthcare professionals

and repeated until the participant was assumed to comply satisfactorily with the equipment. In the current analysis, the cohort was split into individuals with information on number of smoking pack years (Inter99p)

and individuals without this information (Inter99np).

The Jackson Heart Study (JHS)12 is a prospective population-based study to seek the causes of the high

prevalence of common complex diseases among African Americans in the Jackson, Mississippi metropolitan area. During the baseline examination period (2000-2004) 5,301 self-identified African Americans were

recruited from four sources, including (1) randomly sampled households from a commercial listing; (2) ARIC participants; (3) a structured volunteer sample that was designed to mirror the eligible population; and (4) a

nested family cohort. Spirometry was performed as recommended by the American Thoracic Society13 with a dry rolling seal spirometer (Occupational Marketing, Houston, Tex). Measurements included forced vital capacity (FVC), forced expiratory volume at 1 second (FEV1), 3 seconds, and 6 seconds, and peak flow as well

as flows at 25%, 50%, and 75% of total volume. Both calculated data and raw wave-form data were stored for each forced exhalation performed, and quality scores for the test were developed and attached to each

4

participant’s data. Unrelated participants were between 35 and 84 years old, and members of the family cohort were ≥ 21 years old when consent for genetic testing was obtained and blood was drawn for DNA

extraction. A total of 2,790 JHS participants were genotyped on the Illumina Human Exome BeadChip v1. Participants that are also in the ARIC study were excluded for the current analysis.

The KORA studies (Cooperative Health Research in the Region of Augsburg) are a series of independent population based studies from the general population living in the region of Augsburg, Southern Germany14,15.

KORA F4 including 3,080 individuals was conducted from 2006-2008 as a follow-up study to KORA S4 (1999-2001). Lung function tests were performed in a random subsample of subjects born between 1946 and 1965

(age range 41–63 years). Spirometry was performed in line with the ATS/ERS recommendations16 using a pneumotachograph-type spirometer (Masterscreen PC, CardinalHealth, Würzburg, Germany) before and after

inhalation of 200μg salbutamol. The present study is based on maximum values of FEV1 and FVC measured before bronchodilation. The spirometer was calibrated daily using a calibration pump (CardinalHealth,

Würzburg, Germany), and additionally, an internal control was used to ensure constant instrumental conditions. For KORA F4 participants without spirometry measurements in 2006-2008 (n=126), we used

measurements from the KORA-Age time point conducted in 2008/09. KORA Age contains subjects from all KORA studies born until 1943 (aged 65-90 years)17. Spirometry was measured in 935 randomly selected

participants. Conditions including the examiner were the same as in 2008/09 except that inhalation of salbutamol was not performed due to the high number of contraindications anticipated in this aged

population.

The Lothian Birth Cohort 1936 (LBC1936) consists of 1,091 relatively healthy individuals assessed on cognitive and medical traits at about 70 years of age. They were all born in 1936 and most took part in the Scottish

Mental Survey of 1947. At baseline the sample of 548 men and 543 women had a mean age 69.6 years (s.d. = 0.8). They were all Caucasian, community-dwelling, and almost all lived in the Lothian region (Edinburgh city

and surrounding area) of Scotland. A full description of participant recruitment and testing can be found elsewhere18. Genotyping was performed at the Wellcome Trust Clinical Research Facility, Edinburgh. Quality

control measures were applied and 983 participants remained. Lung function assessing peak expiratory flow rate, forced expiratory volume in 1 second, and forced vital capacity (each the best of three), using a Micro

Medical Spirometer was assessed, sitting down without nose clips, at age 70 years. The accuracy of the spirometer is ±3% (to ATS recommendations Standardisation of Spirometry 1994 update for flows and

volumes).

The Multi-Ethnic Study of Atherosclerosis (MESA) is an NHLBI-sponsored population-based longitudinal investigation of subclinical cardiovascular disease and its progression.19 In brief, a total of 6,814 individuals,

aged 45 to 84 years at baseline, were recruited from six US communities between 2000 and August 2002. Thirty-eight percent of the originally recruited participants were White, 28% African-American, 22% Hispanic,

and 12% Asian (predominantly of Chinese descent). Five in-person follow-up examinations have occurred since the MESA cohort was established. The MESA Air Pollution Study recruited an additional 257 participants.20

Spirometry was conducted in 2004-06 in accordance with the American Thoracic Society/European Respiratory Society guidelines21 on a dry-rolling-sealed spirometer with automated quality checks (Occupational

Marketing, Inc., Houston, TX), as previously described,22 for 3,975 participants. Pack-years of cigarette smoking were calculated as age of starting to quitting (or current age if current cigarette smoker) × (cigarettes

per day/20) using standardized questionnaire items. Ever-smoking was defined as greater than 100 lifetime cigarettes smoked and current smoking as self-report of a cigarette in the last 30 days. Current smoking status

was confirmed by cotinine levels in the subset of the cohort with spirometry measures; self-report was generally accurate.23 Asthma was defined as self-report of physician-diagnosed asthma. Height was measured

to the nearest 0.1 cm with the subject in stocking feet and weight was measured to the nearest pound with the subject in light clothing using a balanced scale.

5

The Netherlands Epidemiology of Obesity (NEO) study was designed for extensive phenotyping to investigate pathways that lead to obesity-related diseases. The NEO study is a population-based, prospective cohort study

that includes 6,671 individuals aged 45–65 years, with an oversampling of individuals with overweight or obesity. At baseline, information on demography, lifestyle, and medical history have been collected by

questionnaires. In addition, samples of 24-h urine, fasting and postprandial blood plasma and serum, and DNA were collected. Genotyping was performed using the Illumina HumanCoreExome chip, which was

subsequently imputed to the 1000 genome reference panel. Participants underwent an extensive physical examination, including anthropometry, electrocardiography, spirometry, and measurement of the carotid

artery intima-media thickness by ultrasonography. In random subsamples of participants, magnetic resonance imaging of abdominal fat, pulse wave velocity of the aorta, heart, and brain, magnetic resonance spectroscopy

of the liver, indirect calorimetry, dual energy X-ray absorptiometry, or accelerometer measurements were performed. The collection of data started in September 2008 and completed at the end of September 2012.

Participants are currently being followed for the incidence of obesity-related diseases and mortality.

The Northern Sweden Population Health Study (NSPHS) represents a cross-sectional study conducted in the

communities of Karesuando (samples gathered in 2006) and Soppero (2009) in the subarctic region of the County of Norrbotten, Sweden. Spirometry was performed in sitting position without noseclips using a

MicroMedicalSpida 5 spirometer (http://www. medisave.co.uk). Three consecutive lung function measurements per participant were done and the maximum value per measured lung function parameter was

used for further analysis.

The Prospective Investigation of the Vasculature in Uppsala Seniors (PIVUS)24 is a population-based study of the cardiovascular health in the elderly. The main purpose of PIVUS was to investigate the role endothelial

function in cardiovascular risk. Mailed invitations were sent to subjects who lived in Uppsala, Sweden, within 2 months after their 70th birthday. The subjects were randomly selected from the community register. A total of

1,016 men and women participated in the baseline investigation (participation rate, 50.1%). Spirometry was performed in 901 subjects at baseline in accordance with American Thoracic Society recommendations (α

spirometer; Vitalograph Ltd; Buckingham, UK). The best value from three recordings was used. The Ethics Committee of the University of Uppsala approved the study, and the participants gave their informed consent.

The Rotterdam Study (RS) is an ongoing prospective population-based cohort study, focused on chronic

disabling conditions of the elderly. The study comprises an outbred ethnically homogenous population of Dutch Caucasian origin. The rationale of the study has been described in detail elsewhere.25 In summary, 7,983

men and women aged 55 years or older, living in Ommoord, a suburb of Rotterdam, the Netherlands, were invited to participate. Spirometry was performed from 2009 onwards using a Master Screen® PFT Pro

(CareFusion, San Diego, CA) by trained paramedical personnel according to the ATS/ERS guidelines.26 A total of 546 individuals from the initial study with validated lung function measurement and exome chip data were

included in the current study.

The SAPALDIA cohort is a population-based multi-center study in eight geographic areas representing the range of environmental, meteorological and socio-demographic conditions in Switzerland.27,28 It was initiated

in 1991 (SAPALDIA 1) with a follow-up assessment in 2002 (SAPALDIA 2) and 2010 (SAPALDIA3). This study has specifically been designed to investigate longitudinally lung function, respiratory and cardiovascular health; to

study and identify the associations of these health indicators with individual long term exposure to air pollution, other toxic inhalants, life style and molecular factors.

The Study of Health in West Pomerania (SHIP) is a cross-sectional, population based survey in a region in the Northeast of Germany. Study details are given elsewhere29,30. The examinations were conducted using a bodyplethysmograph equipped with a pneumotachograph (VIASYS Healthcare, JAEGER, Hoechberg, Germany)

which meets the American Thoracic Society (ATS) criteria31. The volume signal of the equipment was calibrated with a 3.0 litre syringe connected to the pneumotachograph in accordance with the manufacturer´s

6

recommendations and at least once on each day´s testing. Barometric pressure, temperature and relative humidity were registered every morning. Calibration of reference gas and volume was examined under ATS-

conditions (Ambient Temperature Pressure) and the integrated volumes were BTPS (Body Temperature Pressure Saturated) corrected31,32. Lung function variables were measured continuously throughout the

baseline breathing and the forced manoeuvres using a VIASYS HEALTHCARE system (MasterScreen Body/Diff.). Spirometry flow volume loops were conducted in accordance with ATS recommendations in a sitting position

and with wearing noseclips. The participants performed at least three forced expiratory lung function manoeuvres in order to obtain a minimum of two acceptable and reproducible values. Immediate on-screen

error codes indicating the major acceptability (including start, duration and end of test) and reproducibility criteria supported the attempt for standardised procedures. The procedure was continuously monitored by a

physician. The best results for FVC, FEV1, peak expiratory flow (PEF) and expiratory flow at 75%, 50%, 25% of FVC (MEF 75, MEF 50, MEF 25) were taken. The ratio of FEV1 to FVC was calculated from the largest FEV1 and

FVC.

The UK Biobank samples comprised of 48,930 individuals selected for the UK Biobank Lung Exome Variant

Evaluation (UK BiLEVE) project33 and a further 49,727 samples selected randomly from UK Biobank, whom had high quality spirometry data.

UK Biobank (http://www.ukbiobank.ac.uk/) contains data from 502,682 individuals (94% of self-reported European ancestry) with extensive health and lifestyle questionnaire data, physical measures (including

spirometry) and DNA. Spirometry was undertaken using a Vitalograph Pneumotrac 6800. The participant was asked to record two to three blows (lasting for at least 6 seconds) within a period of about 6 minutes. The

computer compared the reproducibility of the first two blows and, if acceptable (defined as a <5% difference in forced volume vital capacity (FVC) and Forced Expiratory Volume in 1 second (FEV1), a third blow was not required. For the UK BiLEVE project, a sampling frame of 275,939 individuals was defined as those who were of

European ancestry and had spirometry measures which met ERS/ATS guidelines 16. A total of 50,008 samples were selected from the extremes and middle of the distributions of percent predicted FEV1, separately in never

smokers and heavy smokers (20,005 individuals with low FEV1, 19,997 with average FEV1 and 10,006 with high FEV1). DNA was extracted and genotyped with the custom-designed Affymetrix Axiom UK BiLEVE array. A

further 102,757 individuals from UK Biobank, chosen at random, have subsequently been genotyped using the Affymetrix Axiom UK Biobank array; from these we selected 51,117 individuals of European ancestry and with

spirometry meeting ATS/ERS guidelines. All UK BiLEVE and UK Biobank individuals were imputed to the 1000 Genomes Project Phase 145 and UK10K46,47 combined panel and following quality control of the imputed

genotype data and removal of related individuals, a total of 98,657 individuals remained for association analysis.

The United Kingdom Household Longitudinal Study (UKHLS), also known as Understanding Society (https://www.understandingsociety.ac.uk) is a longitudinal panel survey of 40,000 UK households (England,

Scotland, Wales and Northern Ireland) representative of the UK population. Participants are surveyed annually since 2009 and contribute information relating to their socioeconomic circumstances, attitudes, and

behaviours via a computer assisted interview. The study includes phenotypical data for a representative sample of participants for a wide range of social and economic indicators as well as a biological sample

collection encompassing biometric, physiological, biochemical, and haematological measurements and self-reported medical history and medication use. The United Kingdom Household Longitudinal Study has been

approved by the University of Essex Ethics Committee and informed consent was obtained from every participant. For a subset of individuals who took part in a nurse health assessment, blood samples were taken and genomic

DNA extracted. Of these, 10,484 samples were genotyped at the Wellcome Trust Sanger Institute using the Illumina Infinium HumanCoreExome-12 v1.0BeadChip.

Lung function measures in samples from England and Wales were conducted with the NDD Easy On-PC spirometer (NDD Medical Technologies, Zurich, Switzerland). Participants were excluded in the following cases:

7

pregnancy, having had abdominal or chest surgery (past 3 weeks), admitted to the hospital with a heart complaint (in the past 6 weeks), having had recent eye surgery (past 4 weeks), or in case of having a

tracheostomy. Subjects were asked to perform up to 8 blows that ideally lasted at least 6 seconds, uninterrupted by coughing, glottis closure, laughing or leakage of air. Upon completion, the measurements

were rated either acceptable or unacceptable by the NDD Easy On-PC software.

The Young Finns Study (YFS) is a population-based follow up-study started in 198034. The main aim of the YFS

is to determine the contribution made by childhood lifestyle, biological and psychological measures to the risk of cardiovascular diseases in adulthood. In 1980, over 3,500 children and adolescents all around Finland

participated in the baseline study. The follow-up studies have been conducted mainly with 3-year intervals. The latest 30-year follow-up study was conducted in 2010-2011 (ages 33-49 years) with 2,063 participants. The

study was approved by the local ethics committees (University Hospitals of Helsinki, Turku, Tampere, Kuopio and Oulu) and was conducted following the guidelines of the Declaration of Helsinki. All participants gave their

written informed consent.

Supplementary Methods

Study level Quality Control Procedures For B58C, KORA F4, NSPHS, PIVUS, SAPALDIA, SHIP, FIN and YFS, genotype calling and quality control were

carried out in accordance with the Exome Chip Quality Control SOP Version 5, as developed within the UK exome chip consortium.35 Genotypes were initially called using Gencall in Illumina’s Genome Studio software.36

Quality control of SNPs and samples was subsequently performed at study level. Initial filters applied excluded SNPs with very low call rate (<90%) and samples with low call rate, heterozygosity outliers, duplicates, gender

mismatches and ancestral outliers. SNPs with missing data were then recalled using genotype calling software zCall (zCall not implemented in YFS).37 All alleles were mapped to the forward strand of human genome build

37 and secondary exclusions were applied to remove SNPs with low call rate (<99%) or deviations from Hardy Weinberg Equilibrium (P<10-4). Samples with call rate <99% and heterozygosity outliers were also excluded.

For AGES; ARIC; CHS; FAMHS; FHS; HABC; Health2006; Health 2008; Inter99np; Inter99p; JHS; MESA; RS, GS:SFHS, KORCULA and LBC1936, genotypes were called using Gencall in Illumina’s Genome Studio software36

via the CHARGE Consortium joint calling cluster file (http://www.chargeconsortium.com/main/exomechip) and quality control of the genotype data was undertaken according to the CHARGE exome chip best practices,

described elsewhere.38

UK Biobank: The UK Biobank samples were genotyped using the Affymetrix UK BiLEVE and UK Biobank arrays, which include variants which were selected from the same sequencing project as the Illumina Human Exome

BeadChip and expected to be polymorphic in UK populations, alongside additional content. The QC and imputation procedure of the UK BiLEVE/UK Biobank genotype data is described elsewhere.33,39 In brief,

thorough sample and genotype QC was undertaken before imputation to a combined 1000G40 and UK10K Project41 reference panel. Following imputation, SNPs were excluded if they had imputation INFO score ≤0.5 or

minor allele count (MAC)<3.

UKHLS: Genotype calling was performed using the Illumina GenCall software (2). Sample-level quality control (QC) was performed using the following filters: call rate < 98%, autosomal heterozygosity outliers (> 3 SD),

gender mismatches, duplicates as established by identity by descent (IBD) analysis (PI_HAT > 0.9), ethnic outliers as determined by combining with 1000 Genomes Project data and carrying out IBD followed by

multidimensional scaling. In total, 9,965 samples passed QC. Variant-level QC was performed as follows:

8

variants were mapped to forward strand of human genome build 37. Variants with Hardy-Weinberg equilibrium P < 1×10-4, a call rate < 98% and poor genotype clustering values (< 0.4) were removed, as well as

Y-chromosome and mitochondrial variants.

eQTL Analyses The sentinel SNPs, and proxies (r2>0.8) within the newly identified regions were assessed in the following eQTL

data sets.

Blood: We searched for blood eQTLs within a publicly available blood eQTL dataset with results from the

analysis of 5,311 individuals, imputed to HapMap 2 42. Association testing was undertaken both for cis (+/- 250Kb distance between the SNP and the probe midpoint) and trans (distance between the SNP and the probe

midpoint >5Mb).

GTEx (all tissues): SNPs were assessed in expression data from samples from the GTEx project 43. Only cis-eQTLs (+/- 1Mb distance between the SNP and transcription start site were available in the dataset. All

available tissues and sample sizes are summarised below.

Tissue Number of RNASeq and Genotyped samples

Number of eGenes

Muscle - Skeletal 361 7082

Whole Blood 338 6784

Skin - Sun Exposed (Lower leg) 302 8567

Adipose - Subcutaneous 298 8500

Artery - Tibial 285 8056

Thyroid 278 9937

Lung 278 7236

Nerve - Tibial 256 9860

Esophagus - Mucosa 241 7416

Cells - Transformed fibroblasts 272 8760

Skin - Not Sun Exposed (Suprapubic) 196 5491

Esophagus - Muscularis 218 6916

Adipose - Visceral (Omentum) 185 4301

Artery - Aorta 197 6220

Heart - Left Ventricle 190 4417

Breast - Mammary Tissue 183 4140

Colon - Transverse 169 4446

Heart - Atrial Appendage 159 3929

Stomach 170 3438

Testis 157 9009

Pancreas 149 4301

Esophagus - Gastroesophageal Junction 127 2751

Colon - Sigmoid 124 2882

Adrenal Gland 126 3259

Artery - Coronary 118 2363

Brain - Cerebellum 103 4163

Liver 97 1628

9

Tissue Number of RNASeq and Genotyped samples

Number of eGenes

Cells - EBV-transformed lymphocytes 114 2957

Brain - Caudate (basal ganglia) 100 2447

Brain - Cortex 96 2567

Brain - Nucleus accumbens (basal ganglia) 93 2019

Brain - Frontal Cortex (BA9) 92 2009

Prostate 87 1462

Brain - Cerebellar Hemisphere 89 3251

Spleen 89 2754

Pituitary 87 2160

Brain - Putamen (basal ganglia) 82 1588

Ovary 85 1583

Brain - Hypothalamus 81 1157

Vagina 79 840

Brain - Hippocampus 81 1134

Small Intestine - Terminal Ileum 77 1356

Brain - Anterior cingulate cortex (BA24) 72 1212

Uterus 70 917

Lung resection cohort: Selected SNPs were assessed in a lung eQTL resource based on lung tissues of 1,111

individuals. The descriptions of the lung eQTL dataset and subject demographics have been published previously 44–46. Briefly, non-tumor lung tissues were collected from patients who underwent lung resection

surgery at three participating sites: Laval University (Quebec City, Canada), University of Groningen (Groningen, The Netherlands), and University of British Columbia (Vancouver, Canada). Whole-genome gene

expression and genotyping data were obtained from these specimens. Gene expression profiling was performed using an Affymetrix custom array (GPL10379) testing 51,627 non-control probe sets and normalized

using RMA47. Genotyping was performed using the Illumina Human1M-Duo BeadChip array. Genotype imputation was undertaken using the 1000 Genomes Project 40 reference panel. Following standard microarray

and genotyping quality controls, 1111 patients were available including 409 from Laval, 363 from Groningen, and 339 from UBC. Lung eQTLs were identified to associate with mRNA expression in either cis (within 1 Mb of

transcript start site) or in trans (all other eQTLs).

Analysis: Look-ups in the lung resection cohort, blood and GTEx (all tissues) datasets were undertaken for the 7 sentinel SNPs identified in the combined SpiroMeta-CHARGE analysis, and all proxies with r2>0.8. For all

datasets, eQTL associations were identified as significant at an FDR of 5% (or q-value<0.05). For any gene identified in the eQTL look-up, the most significantly associated eSNP for each gene was identified. All genes

for which the sentinel lung function SNP and the top eSNP were in strong linkage disequilibrium (r2>0.9) were further highlighted (i.e. where there was co-localisation of the lung function associated SNP and the gene

expression associated SNP).

10

Supplementary Figures Supplementary Figure1

Quantile-quantile (QQ) and Manhattan plots for consortium-wide analyses (CHARGE, top; SpiroMeta, middle), and the combined meta-

analysis (bottom). Results shown are for the analysis of all individuals (never and ever smokers combined, and both European and African

ancestry), for three lung function traits: A. FEV1, B. FVC C. FEV1/FVC.

A. FEV1

A.

11

B. FVC

12

C. FEV1/FVC

13

Supplementary Figure 2: Region Plots for novel loci.

14

15

16

17

Supplementary Figure 3: Forest plots of novel loci.

18

19

Supplementary Figure 4: Trait Transformation Sensitivity Analysis. We have repeated the analysis for the six reported SNPs (the LCT SNP was not available in UK Biobank), transforming the phenotypes, and then adjusting for all covariates (including PCs)

during the SNP-trait association test. For comparison, we have done this for all six SNPs with all three traits. Comparisons of these two analyses (not adjusted prior to transformation vs with

adjustment prior to transformation) are shown here. For each SNP, the P-value comparison is highlighted for the trait we report the association with, and the dashed lines indicate the

Bonferroni corrected significance threshold for independent replication (P<1·47×10 -3). Whilst there is a difference in the P-values for some SNP-trait combinations, (more significant P-values

in the analysis with covariate adjustment prior to transformation for 5 of the 6 SNPs), the SNPs all meet the replication P-value threshold in both analyses.

20

Supplementary Tables Supplementary Table 1: Details of study specific genotyping platform, genotype calling procedure and software.

Discovery Cohorts Study Name Genotyping Platform Calling algorithm(s) Study-level software B58C Illumina Human Exome BeadChip v1 Gencall + zCall RAREMETALWORKER

GS:SFHS Illumina Human Exome BeadChip v1 CHARGE RAREMETALWORKER

KORA F4 Illumina Human Exome BeadChip v1 Gencall + zCall RAREMETALWORKER

KORCULA Illumina Human Exome BeadChip v1 CHARGE RAREMETALWORKER

LBC 1936 Illumina Human Exome BeadChip v1 CHARGE rvtests

SHIP Illumina Human Exome BeadChip v1 Gencall + zCall RAREMETALWORKER

NSPHS Illumina Human Exome BeadChip v1 Gencall + zCall RAREMETALWORKER

PIVUS Illumina Human Exome BeadChip v1 Gencall + zCall RAREMETALWORKER

SAPALDIA Illumina Human Exome BeadChip v1 Gencall + zCall RAREMETALWORKER

YFS Illumina Human Exome BeadChip v1 Gencall rvtests

FIN Illumina Human Exome BeadChip v1 Gencall + zCall RAREMETALWORKER

AGES Illumina Human Exome BeadChip v1 CHARGE Seqmeta

ARIC Illumina Human Exome BeadChip v1 CHARGE Seqmeta

CHS Illumina Human Exome BeadChip v1 CHARGE Seqmeta

FAMHS Illumina Human Exome BeadChip v1 CHARGE Seqmeta

FHS Illumina Human Exome BeadChip v1 CHARGE Seqmeta

HABC Illumina Human Exome BeadChip v1 CHARGE Seqmeta

Health2006 Illumina Human Exome BeadChip v1 CHARGE Seqmeta

Health2008 Illumina Human Exome BeadChip v1 CHARGE Seqmeta

Inter99np Illumina Human Exome BeadChip v1 CHARGE Seqmeta

Inter99p Illumina Human Exome BeadChip v1 CHARGE Seqmeta

JHS Illumina Human Exome BeadChip v1 CHARGE Seqmeta

MESA Illumina Human Exome BeadChip v1 CHARGE Seqmeta

RS Illumina Human Exome BeadChip v1 CHARGE Seqmeta

Replication Cohorts Study Name Genotyping Platform Calling algorithm(s) Study-level software

UK Biobank Affymetrix Axiom UK BiLEVE array and Affymetrix Axiom UK Biobank array

Axiom® GT1 algorithm (Affymetrix Power Tools v1.15.1)

Imputation: SHAPEIT and IMPUTE2. Association Testing: SNPTEST (single variant associations, imputed data); RAREMETALWORKER (gene-based associations, genotyped SNPs only).

UKHLS Illumina Infinium HumanCoreExome-12 v1.0BeadChip GenCall RAREMETALWORKER

NEO Illumina Infinium HumanCoreExome-12 v1.0BeadChip CHARGE Seqmeta

21

Supplementary Table 2: Association results for all SNPs identified in single variant association discovery analyses (P<10-4). Only variants in novel loci shown, and results presented for the trait, smoking and ancestry combination for which the strongest association was identified. Variants were followed up for the trait and smoking subset

for which they were most significantly associated (only European Ancestry samples available). Chromosome (CHROM) and position (POS) in build 37 are given for each SNP. For the SpiroMeta Consortium, beta values reflect effect-size estimates on an inverse-normal transformed scale after adjustments for age, age2, sex, height and ancestry

principal components. For the CHARGE Consortium, beta values represent untransformed trait effect estimates (ml or ratio), after adjustment for former smoking, current smoking and pack-years of smoking, age, age2, sex, height , height2, centre/cohort, principle components and weight (FVC only).

Consortium level Discovery Analysis

CHARGE Consortium SpiroMeta Consortium Discovery Meta-analysis Overall Replication

Combined Discovery + Replication Result

dbSNPID Chr:Pos

effect/ noneffect allele

(nearest) gene Trait

Smoking subset (ever /never smokers / all)

Ancestry subset (both [EA+AA] / EA only) Beta P-value Beta P-value N

Effect Allele Frequency (EAF) Z-score P-value N EAF Z-score P-value Z-score P-value

rs61744357 1:16134072 T/C UQCRHL FVC all both -17.522 1.04E-3 -0.070 3.58E-4 57363 15.092% -4.516 6.29E-6 104113 11.641% 0.455 0.6489 -2.326 2.00E-2

rs41273537 1:150429944 G/A RPRD2 FEV1 never both -30.677 1.39E-4 -0.072 2.96E-4 28519 10.789% -5.197 2.03E-7 47087 12.019% -1.754 0.0795 -4.552 5.32E-6

rs202079239 1:154548277 G/C CHRNB2 FEV1 ever EA 909.281 6.21E-5 0.998 3.69E-3 27269 0.026% 4.885 1.03E-6 4506 0.022% 1.437 0.1507 5.066 4.05E-7

rs145942857 2:45801809 C/T SRBD1 FEV1/FVC all EA -9.948 1.80E-5 -0.604 2.69E-2 60381 0.021% -4.650 3.32E-6 7443 0.034% 0.765 0.4442 -4.062 4.87E-5

rs2322659 2:136555659 T/C LCT FVC all EA 27.341 2.82E-7 0.032 7.60E-3 55591 23.560% 5.597 2.18E-8 12899 20.134% 2.286 0.0223 6.024 1.70E-9

rs147184138 3:25833094 G/C OXSM FVC ever EA 321.888 7.05E-3 0.903 7.53E-5 27268 0.077% 4.609 4.04E-6 4506 0.078% 0.253 0.8004 4.365 1.27E-5

rs201776558 3:54922021 A/G CACNA2D3 FVC ever EA -1196.294 3.15E-4 -1.393 2.39E-3 27268 0.015% -4.698 2.63E-6 4504 0.000% 0.000 1.0000 -4.352 1.35E-5

rs150210060 3:113753937 A/G KIAA1407 FVC ever EA 135.506 4.08E-5 0.156 2.19E-2 27268 0.942% 4.592 4.38E-6 64472 1.102% -0.333 0.7394 2.225 2.61E-2

rs35706839 3:182733255 T/C MCCC1 FVC all EA 1159.392 5.02E-5 2.054 5.16E-3 60394 0.005% 4.861 1.17E-6 7441 0.000% 0.000 1.0000 4.534 5.78E-6

rs199529766 3:185783622 T/C ETV5 FEV1 ever EA -1346.179 2.41E-4 -2.768 5.77E-3 27269 0.006% -4.542 5.57E-6 4506 0.011% 0.458 0.6472 -4.035 5.45E-5

rs202225480 4:10502971 G/A CLNK FEV1 ever EA 1163.190 6.29E-4 1.528 1.02E-3 27269 0.013% 4.690 2.74E-6 4506 0.000% 0.000 1.0000 4.344 1.40E-5

rs1448044 5:44296986 A/G FGF10 FVC ever both 18.628 2.02E-3 0.056 8.48E-5 30966 35.637% 4.813 1.49E-6 64400 32.646% 4.805 1.55E-6 6.691 2.22E-11

rs1294421 6:6743149 T/G LY86 FEV1/FVC all both -0.222 3.83E-5 -0.037 1.22E-4 68099 36.786% -5.479 4.27E-8 111556 39.273% -8.171 3.06E-16 -9.815 9.74E-23

rs200718368 6:28053978 C/T ZNF165 FEV1/FVC never both -14.186 1.55E-6 -0.685 4.08E-2 28504 0.025% -4.953 7.29E-7 2940 0.000% 0.000 1 -4.603 4.17E-6

rs4713479 6:31688799 T/C LY6G6C FEV1/FVC ever both -0.799 1.24E-4 -0.107 9.96E-3 30967 4.250% -4.551 5.33E-6 60910 2.974% -1.710 0.0873 -4.035 5.47E-5

22

Consortium level Discovery Analysis

CHARGE Consortium SpiroMeta Consortium Discovery Meta-analysis Overall Replication

Combined Discovery + Replication Result

dbSNPID Chr:Pos

effect/ noneffect allele

(nearest) gene Trait

Smoking subset (ever /never smokers / all)

Ancestry subset (both [EA+AA] / EA only) Beta P-value Beta P-value N

Effect Allele Frequency (EAF) Z-score P-value N EAF Z-score P-value Z-score P-value

rs200840680 7:138978128 C/T UBN2 FEV1/FVC never both -10.334 5.83E-5 -1.043 2.25E-2 28504 0.018% -4.510 6.49E-6 2937 0.000% 0.000 1 -4.191 2.78E-5

rs143386950 8:69033248 T/C PREX2 FEV1/FVC ever both -47.934 2.64E-5 -2.187 2.90E-2 30967 0.003% -4.598 4.26E-6 4506 0.000% 0.000 1 -4.296 1.74E-5

rs201821244 8:144808545 G/T FAM83H FEV1/FVC all EA -29.719 3.90E-5 -2.911 4.16E-3 60381 0.002% -4.914 8.92E-7 7441 0.000% 0.000 1 -4.584 4.56E-6

rs189491808 9:79321371 A/G PRUNE2 FEV1/FVC ever EA -7.332 2.15E-4 -0.560 9.40E-3 27260 0.073% -4.448 8.65E-6 8068 0.087% -0.997 0.3186 -4.384 1.16E-5

rs3849969 10:75525999 T/C SEC24C FEV1 all both 13.095 3.83E-4 0.036 6.99E-4 68116 29.402% 4.767 1.87E-6 111556 27.835% 5.042 4.60E-7 6.906 4.99E-12

rs148545029 11:102589238 A/G MMP8 FEV1 never both -1168.532 1.88E-6 -0.766 4.46E-2 28519 0.018% -4.893 9.94E-7 2940 0.078% 0.549 0.5831 -4.344 1.40E-5

rs149662873 12:7015117 G/A LRRC23 FEV1/FVC never both -19.289 1.17E-6 -1.759 1.29E-2 28504 0.007% -5.279 1.30E-7 2940 0.011% 1.036 0.3001 -4.523 6.11E-6

rs142171275 14:57741119 G/A MUDENG FVC never both -575.432 4.11E-4 -1.174 2.14E-3 28515 0.030% -4.612 3.99E-6 2940 0.011% -0.431 0.6665 -4.445 8.78E-6

rs1200345 15:41819716 C/T RPAP1 FEV1/FVC all EA -0.216 9.02E-5 -0.025 9.92E-3 60381 48.785% -4.586 4.51E-6 111556 49.465% -5.725 1.03E-8 -7.329 2.33E-13

rs141062694 16:31049855 T/C STX4 FEV1/FVC ever both -18.063 2.00E-5 -1.661 2.13E-2 30967 0.011% -4.718 2.38E-6 4506 0.011% 0.375 0.7073 -4.274 1.92E-5

rs142933706 17:3921203 C/G ZZEF1 FVC never both -1961.144 4.36E-5 -2.288 2.24E-2 28515 0.004% -4.531 5.86E-6 2940 0.000% 0.000 1 -4.211 2.54E-5

rs144838552 17:7838371 T/C CNTROB FEV1/FVC ever EA -34.085 2.49E-6 -0.944 3.73E-2 26835 0.011% -4.887 1.02E-6 60910 0.001% -1.358 0.1743 -3.834 1.26E-4

rs139231634 17:51901858 A/C KIF2B FEV1/FVC ever both -3.523 2.84E-5 -0.254 4.62E-2 30967 0.252% -4.471 7.80E-6 60910 0.231% -1.532 0.1255 -3.843 1.22E-4

rs1859962 17:69108753 G/T CASC17 FEV1 all EA 15.389 2.75E-5 0.026 7.50E-3 60395 48.171% 4.876 1.08E-6 111554 46.464% 4.612 3.99E-6 6.600 4.10E-11

rs61739354 17:78039309 T/G CCDC40 FEV1/FVC never EA 21.157 1.89E-4 2.660 9.40E-3 24483 0.004% 4.441 8.94E-6 2940 0.000% 0.000 1 4.082 4.47E-5

rs6088813 20:33975181 C/A UQCC1 FVC all both -16.157 3.44E-5 -0.023 2.29E-2 68115 41.441% -4.634 3.58E-6 111556 36.898% -7.688 1.50E-14 -8.915 4.90E-19

rs6004919 22:26455216 T/C NA FVC ever both -37.193 4.82E-5 -0.052 3.31E-2 30966 15.546% -4.488 7.19E-6 8068 8.193% -1.342 0.1795 -4.608 4.07E-6

rs112306225 22:37261097 A/C NCF4 FEV1/FVC ever EA -5.084 9.00E-6 -0.347 1.18E-2 26437 0.193% -4.939 7.85E-7 64472 0.151% -0.675 0.5000 -3.231 1.23E-3

rs147770992 22:46795742 T/C CELSR1 FEV1 never EA -764.555 4.17E-5 -0.809 3.30E-2 24497 0.024% -4.419 9.93E-6 2940 0.011% 0.329 0.7421 -3.931 8.45E-5

23

Supplementary Table 3: Association results for SNPs identified in single variant association discovery analyses (P<10-4), located in known lung function regions. Variants that are in regions previously identified as showing association with lung function traits. For each variant, the result presented is for the trait, smoking and ancestry

combination for which the strongest association was identified.

rs id

Effect / other allele (Nearest) Gene Trait

Smoking subset (ever

/never smokers /

all)

Ancestry subset (both

[EA+AA] / EA only)

Discovery Analysis

Discovery meta-analysis

Previous Associations

SpiroMeta Consortium CHARGE Consortium

Beta P-value Beta P-value N

Effect Allele Frequency (EAF) Z-score P-value

rs1192415 1:92077097 G/A - FEV1/FVC all both -0.044 2.99E-4 -0.204 2.34E-3 68099 18.1% -4.497 6.91E-6 rs1192404 (r2=0.951) - Associated with FEV1/FVC39

rs2571445 2:218683154 A/G TNS1 FEV1 all both -0.032 1.16E-3 -16.757 2.11E-6 68116 36.8% -5.631 1.79E-8 SNP associated with FEV1 48

rs12477314 2:239877148 T/C - FEV1/FVC all both 0.034 3.77E-3 0.318 2.42E-6 68099 18.6% 5.411 6.26E-8 SNP associated with FEV1/FVC 48

rs7671167 4:89883979 C/T FAM13A FEV1/FVC all both 0.043 5.87E-6 0.307 3.60E-9 68099 48.8% 7.293 3.03E-13 COPD locus; rs2045517 (r2=0.664) - associated with FEV1/FVC 48,49

rs2280099 4:90035549 G/A TIGD2 FEV1/FVC all EA -0.059 1.63E-6 -0.183 0.012 60381 17.8% -4.870 1.12E-6 rs7671167 (r2=0.150 ), Cho FAM13A and rs2045517 (r2=0.192) - associated with FEV1/FVC 48

rs10516526 4:106688904 G/A GSTCD FEV1 all EA 0.076 9.41E-5 44.087 3.79E-9 60395 6.4% 6.943 3.83E-12 SNP previously associated with FEV150

rs13118928 4:145486389 G/A - FEV1/FVC all both 0.034 4.08E-4 0.407 7.41E-14 68099 37.9% 7.975 1.53E-15 rs138641402 (r2>0.9) - associated with FEV1/FVC 48

rs11168048 5:147842353 C/T HTR4 FEV1/FVC all both 0.045 3.92E-6 0.405 3.51E-14 68099 39.3% 8.673 4.19E-18 rs1985524 (r2=1 ) associated with FEV1 48

rs1422795 5:156936364 C/T ADAM19 FEV1/FVC all both -0.037 2.35E-4 -0.362 1.76E-11 68099 36.7% -7.457 8.86E-14 SNP associated with FEV1/FVC 48

rs2070600 6:32151443 T/C AGER FEV1/FVC all both 0.049 2.56E-2 0.900 2.33E-11 68099 4.6% 6.593 4.31E-11 SNP associated with FEV1/FVC 48

rs17280293 6:142688969 G/A GPR126 FEV1/FVC all both 0.125 3.29E-5 0.962 6.72E-9 68099 2.4% 6.993 2.69E-12 rs148274477 (r2=0.85) associated with FEV1/FVC 51

rs11155242 6:142691549 C/A GPR126 FEV1/FVC all EA 0.040 6.66E-4 0.503 1.17E-12 60381 19.2% 7.545 4.51E-14 rs262129 (r2=0.522 ) - associated with FEV1/FVC 48

rs16909898 9:98231008 G/A LOC100507346 FEV1/FVC all EA -0.041 9.99E-3 -0.448 1.19E-6 60381 9.9% -5.309 1.10E-7 SNP associated with FEV1/FVC 48

rs12684650 9:139110654 T/C QSOX2 FVC all both -0.032 2.00E-3 -16.905 1.22E-4 68115 28.1% -4.866 1.14E-6 rs10858246 (r2=0.887) - associated with FVC 51

rs7068966 10:12277992 T/C CDC123 FEV1/FVC all both 0.027 4.79E-3 0.295 1.77E-8 68099 48.9% 6.097 1.08E-9 rs7090277(r2=1) - associated with FEV1/FVC39

rs241 10:78444456 G/T - FEV1 never both 0.029 0.027 19.154 4.15E-5 28519 50.0% 4.517 6.26E-6 rs2637254 (r2=0.657) - associated with FEV1 48

rs11555762 11:43876698 T/C HSD17B12 FVC all EA -0.036 5.89E-4 -16.963 1.80E-4 60394 30.3% -5.015 5.30E-7 rs4237643 (r2=0.594) - associated with FVC 52

rs10843206 12:28722756 C/T - FVC all EA 0.041 2.09E-5 18.870 6.80E-6 60394 50.5% 6.103 1.04E-9 rs2348418 (r2=0.853)- associated with FVC 51

rs11172113 12:57527283 C/T LRP1 FEV1/FVC never both 0.033 1.32E-2 0.305 2.61E-5 28504 40.7% 4.776 1.79E-6 SNP associated with FEV1/FVC 48

rs2173958 12:115179246 G/T TBX3 FEV1 all both -0.031 1.58E-3 -12.325 2.72E-4 68116 42.5% -4.705 2.53E-6 rs10850377 (r2=0.273)- associated with FEV1 51

rs10498635 14:93103309 T/C RIN3 FEV1 all both 0.037 2.93E-3 20.912 3.20E-6 68116 17.0% 5.407 6.42E-8 rs117068593 (r2=0.991) - associated with FEV148

24

rs id

Effect / other allele (Nearest) Gene Trait

Smoking subset (ever

/never smokers /

all)

Ancestry subset (both

[EA+AA] / EA only)

Discovery Analysis

Discovery meta-analysis

Previous Associations

SpiroMeta Consortium CHARGE Consortium

Beta P-value Beta P-value N

Effect Allele Frequency (EAF) Z-score P-value

rs12899618 15:71645120 A/G THSD4 FEV1/FVC all EA -0.063 1.85E-6 -0.420 4.90E-8 60381 15.3% -7.109 1.17E-12 rs10851839 (r2=0.349) associated with FEV1/FVC 50

rs8034191 15:78806023 C/T AGPHD1 FEV1 ever EA -0.064 6.64E-6 -29.287 1.19E-6 27269 34.1% -6.559 5.40E-11 Smoking/COPD locus 53

rs12447804 16:58075282 T/C MMP15 FEV1/FVC all both -0.019 0.105 -0.300 6.94E-6 68099 19.3% -4.502 6.72E-6 SNP associated with FEV1/FVC 48

rs34093919 19:41117300 A/G LTBP4 FEV1/FVC all EA 0.191 2.24E-5 0.656 5.57E-3 60381 1.3% 4.723 2.33E-6 rs113473882 (r2=0.878), associated with FEV1/FVC.51

25

Supplementary Table 4: Single variant association result for the seven novel signals, in smoking and ancestry subgroups. Smoking subgroups are ever smokers, never smokers and all individuals. Ancestry subgroups are

European Ancestry only (EA) and both ancestries combined (European and African ancestries).

SNP Trait Ancestry Smoking Analysis

CHARGE Beta

CHARGE P-value

SpiroMeta Beta

SpiroMeta P-value

Meta Z-score

Meta P-value

rs2322659 FVC EA ever 22.632 4.67E-3 0.036 0.029 3.550 3.85E-4

never 33.311 2.53E-5 0.028 0.089 4.088 4.36E-5

all 27.341 2.82E-7 0.032 7.60E-3 5.597 2.18E-8 both ever 14.976 0.031 0.036 0.029 3.027 2.47E-3

never 25.048 1.66E-4 0.028 0.089 3.889 1.01E-4

all 21.781 2.94E-6 0.032 7.60E-3 5.278 1.30E-7

rs1448044 FVC EA ever 19.713 3.47E-3 0.056 8.48E-5 4.763 1.90E-6

never 6.620 0.323 -0.008 0.591 0.332 0.740

all 10.966 0.014 0.024 0.022 3.317 9.11E-4

both ever 18.628 2.02E-3 0.056 8.48E-5 4.813 1.49E-6 never -0.395 0.946 -0.008 0.591 -0.392 0.695

all 6.963 0.083 0.024 0.022 2.712 6.68E-3

rs1294421 FEV1/FVC EA ever -0.174 0.064 -0.045 9.19E-4 -3.524 4.25E-4

never -0.204 0.013 -0.029 0.028 -3.284 1.02E-3

all -0.203 3.64E-4 -0.037 1.22E-4 -5.089 3.60E-7

both ever -0.197 0.026 -0.045 9.19E-4 -3.741 1.83E-4

never -0.230 2.11E-3 -0.029 0.028 -3.734 1.88E-4

all -0.222 3.83E-5 -0.037 1.22E-4 -5.479 4.27E-8 rs3849969 FEV1 EA ever 18.212 5.40E-3 0.035 0.021 3.584 3.39E-4

never 13.253 0.030 0.038 1.01E-2 3.336 8.49E-4

all 14.486 5.41E-4 0.036 6.99E-4 4.749 2.04E-6

both ever 14.775 9.70E-3 0.035 0.021 3.426 6.12E-4

never 13.527 9.38E-3 0.038 0.010 3.609 3.07E-4

all 13.095 3.83E-4 0.036 6.99E-4 4.767 1.87E-6 rs1200345 FEV1/FVC EA ever -0.340 2.05E-4 -0.031 0.020 -4.279 1.87E-5

never -0.213 7.17E-3 -0.018 0.171 -2.853 4.33E-3

all -0.216 9.02E-5 -0.025 9.92E-3 -4.586 4.51E-6 both ever -0.263 2.00E-3 -0.031 0.020 -3.812 1.38E-4

never -0.216 2.62E-3 -0.018 0.171 -3.158 1.59E-3

all -0.190 2.45E-4 -0.025 9.92E-3 -4.396 1.10E-5

rs1859962 FEV1 EA ever 15.508 6.18E-3 0.040 2.99E-3 3.974 7.06E-5

never 15.789 3.22E-3 0.012 0.370 2.736 6.22E-3

all 15.389 2.75E-5 0.026 7.50E-3 4.876 1.08E-6 both ever 14.594 4.71E-3 0.040 2.99E-3 4.012 6.03E-5

never 11.613 0.016 0.012 0.370 2.403 0.016

all 13.308 7.43E-5 0.026 7.50E-3 4.681 2.85E-6

rs6088813 FVC EA ever -7.325 0.261 -0.017 0.232 -1.627 0.104

never -10.242 0.111 -0.029 0.036 -2.559 0.011

all -13.728 1.35E-3 -0.023 0.023 -3.882 1.04E-4

both ever -10.791 0.067 -0.017 0.232 -2.167 0.030

never -14.142 0.013 -0.029 0.036 -3.207 1.34E-3

all -16.157 3.44E-5 -0.023 0.023 -4.634 3.58E-6

26

Supplementary Table 5: Single variant association result for rs1448044 and FVC in ever smokers and never smokers separately, and in all samples combined. Replication

Result presented includes data from UK Biobank and UKHLS only (result from NEO available only from ever smokers). Original combined discovery and replication result in

ever smokers, including NEO: P= 2.22x10-11.

Consortium level Discovery Analysis

CHARGE Consortium SpiroMeta Consortium Discovery Meta-analysis Overall Replication Combined Discovery +

Replication Result

SNP Trait Ancestry Smoking Analysis Beta P-value Beta P-value Z-score P-value Z-score P-value Z-score P-value

rs1448044 FVC both ever 18.628 2.02E-3 0.056 8.48E-5 4.813 1.49E-6 4.163 3.14E-5 6.354 2.10E-10

never -0.395 0.946 -0.008 0.591 -0.392 0.695 3.809 1.40E-4 2.385 1.71E-2

all 6.963 0.083 0.024 0.022 2.712 6.68E-3 5.638 1.72E-8 5.638 4.22E-9

27

Supplementary Table 6: Association results for all genes identified in discovery SKAT analyses (meta-analysis P<10-4). N SNPs is the number of SNPs included in the SKAT

test. For the two each genes with evidence of replication, a conditional analysis was carried out, conditioning on the most significantly associated individual SNP within that

gene. In each case, the individual SNP id is given, along with conditional SKAT P-value when that SNP has been conditioned on. Only the trait, smoking ancestry

combination for which each gene was most significantly associated is shown.

Discovery Analysis

Replication Analysis CHARGE Consortium SpiroMeta Consortium Meta-

Analysis

GENE Trait

Smoking subset (ever /never smokers / all)

Ancestry subset (both [EA+AA] / EA only) SNP set N SNPs P-value N SNPs P-value P-value N SNPs P-value Notes

GPR126 FEV1/FVC all both exonic 25 9.98E-10 20 4.42E-5 1.40E-12 11 2.03E-40 Driven by rs17280293 (known signal); conditional P=0.287

LTBP4 FEV1/FVC all EA exonic 25 3.87E-3 25 3.49E-5 2.27E-6 17 2.23E-15 Driven by rs34093919 (known signal); conditional P=0.4552

NCF4 FEV1/FVC ever EA exonic 12 8.08E-6 11 4.05E-2 5.21E-6 3 0.805

LY6G6D FEV1/FVC ever both exonic 3 6.60E-5 2 8.83E-3 8.96E-6 NA NA

IL17RB FEV1/FVC all EA LOF 3 1.91E-3 2 5.96E-4 1.67E-5 NA NA

CPVL FEV1 never EA LOF 3 6.79E-5 3 2.90E-2 2.78E-5 NA NA

SEPT12 FVC never both exonic 12 1.59E-4 8 1.54E-2 3.41E-5 3 0.525

GUCA1C FVC ever both exonic 4 3.77E-4 3 7.89E-3 4.08E-5 1 0.087

CC2D2B FEV1 all both exonic 9 2.64E-4 6 1.18E-2 4.27E-5 2 0.800

ALOXE3 FVC never EA exonic 15 1.81E-2 13 1.87E-4 4.60E-5 4 0.599

ZNF491 FEV1/FVC never both LOF 2 9.89E-5 2 4.25E-2 5.63E-5 NA NA

FAS FVC never both exonic 6 1.36E-3 3 4.12E-3 7.34E-5 2 0.026

USP49 FEV1 all both exonic 8 5.77E-4 7 1.15E-2 8.56E-5 2 0.776

ANO6 FEV1/FVC ever both exonic 20 1.46E-2 16 4.73E-4 8.91E-5 6 0.495

RAD18 FEV1/FVC all both exonic 13 3.06E-3 12 2.47E-3 9.66E-5 2 0.776

MXI1 FVC never both exonic 3 1.35E-3 2 5.75E-3 9.93E-5 4 0.726

28

Supplementary Table 7: Association results for all genes identified in discovery Weighted sum test (WST) test analyses (P<10-4). N SNPs is the number of SNPs included in

the WST. Betas values for SpiroMeta Consortium and Replication results reflect effect-size estimates on an inverse-normal transformed scale. For the CHARGE Consortium,

beta values represent untransformed trait effect estimates. Only the trait, smoking ancestry combination for which each gene was most significantly associated is shown.

Discovery Analysis

Replication Analysis CHARGE Consortium SpiroMeta Consortium Meta-

Analysis

GENE Trait

Smoking subset (ever /never smokers / all)

Ancestry subset (both [EA+AA] / EA only) SNP set N SNPs Beta P-value N SNPs Beta P-value P-value N SNPs Beta P-value

EDARADD FEV1/FVC all EA exonic 4 -0.1729 2.77E-5 2 -0.0067 3.21E-2 3.93E-6 1 0.0028 0.2150

AAMP FVC all both exonic 9 -5.1496 6.35E-3 5 -0.0142 6.29E-4 2.48E-5 3 0.0006 0.6604

LY6G6D FEV1 all EA exonic 2 -13.4532 3.56E-4 2 -0.0108 2.44E-2 2.72E-5 NA NA NA

GSTM5 FEV1 never EA exonic 3 -13.9157 3.64E-3 4 -0.0130 3.51E-3 3.79E-5 2 0.0022 0.3585

RCN3 FVC all EA exonic 9 6.1012 2.39E-3 8 0.0108 5.58E-3 4.09E-5 5 0.0013 0.2100

TMPRSS7 FEV1 never both LOF 3 -14.4113 7.11E-4 2 -0.1528 2.19E-2 4.86E-5 NA NA NA

GATA5 FEV1/FVC ever EA exonic 3 -0.2505 1.74E-3 3 -0.0099 1.02E-2 5.16E-5 3 -0.0011 0.5227

CHTF18 FEV1 never both exonic 49 -3.7700 4.43E-4 42 -0.0038 3.20E-2 5.19E-5 15 0.0008 0.3193

SALL2 FEV1/FVC all both exonic 13 -0.0708 2.26E-3 10 -0.0059 7.88E-3 5.48E-5 NA NA NA

SH2D3A FVC all EA LOF 2 13.5166 1.25E-3 2 0.0817 1.56E-2 5.52E-5 NA NA NA

NUP188 FEV1 ever both exonic 36 3.7176 1.93E-3 24 0.0032 1.52E-2 8.24E-5 10 -0.0005 0.5733

RBPMS2 FEV1/FVC all both exonic 2 -0.1670 2.06E-3 2 -0.0108 1.51E-2 8.72E-5 2 -0.0029 0.0657

29

Supplementary Table 8: Evidence for the role of novel variants identified in single variant association analyses as eQTLs For each probeset variants were ranked firstly according to their correlation with the sentinel SNP (r2) and secondly by eQTL P value, and only the top ranked SNP for each probeset is presented (eQTL SNP). Results in bold indicate there is co-localisation of the GWAS and eQTL signal, that is sentinel SNP and the top eSNP for that gene/tissue combination have r2>0.9. Lung† indicates result from the lung resection cohort, Blood‡ indicates the result is from the Westra et al.42 blood dataset; all other results from GTEx. All identified eQTLs are cis.

sentinel Gene eQTL SNP r2 with sentinel tissue P-value

rs1200345 JMJD7-PLA2G4B rs1200349 1.000 Lung† 1.54E-6

rs1200345

LTK

rs11632399 1.000 Lung† 6.90E-65

rs2297378 0.919 Small_Intestine_Terminal_Ileum 3.94E-6

rs7178957 0.942 Artery_Tibial 1.40E-5

rs1200345 MAPKBP1 rs11632399 1.000 Lung† 3.66E-29

rs1200345

NDUFAF1

rs1200345 1.000 Muscle_Skeletal 7.71E-7

rs28860492 0.996 Lung 1.00E-5

rs1200345

RPAP1

rs1200349 1.000 Blood‡ 6.93E-21

rs1200345 1.000 Esophagus_Mucosa 2.33E-17

rs1200349 1.000 Lung† 2.52E-16

rs1200345 1.000 Nerve_Tibial 3.31E-15

rs1200345 1.000 Colon_Transverse 3.37E-13

rs1200345 1.000 Artery_Aorta 2.18E-12

rs1200345 1.000 Adipose_Visceral_Omentum 4.94E-11

rs1200345 1.000 Thyroid 2.47E-10

rs1200345 1.000 Whole_Blood 1.99E-9

rs1200345 1.000 Breast_Mammary_Tissue 2.43E-9

rs1200345 1.000 Adrenal_Gland 1.60E-8

rs1200345 1.000 Esophagus_Gastroesophageal_Junction 5.55E-8

rs1200345 1.000 Brain_Putamen_basal_ganglia 1.30E-7

rs1200345 1.000 Testis 4.71E-7

rs1200345 1.000 Vagina 5.23E-7

rs2297380 0.838 Brain_Frontal_Cortex_BA9 8.06E-6

rs1200345

SPTBN5

rs11632399 1.000 Lung† 7.12E-8

rs1200345 1.000 Adipose_Subcutaneous 1.09E-7

rs1200345 1.000 Lung 2.82E-7

rs1200345 1.000 Cells_Transformed_fibroblasts 2.07E-6

rs1200345 1.000 Adipose_Visceral_Omentum 2.44E-6

rs1200345 1.000 Thyroid 3.50E-6

rs1200345 1.000 Muscle_Skeletal 1.48E-5

rs1200345 1.000 Artery_Aorta 1.67E-5

rs1200345 1.000 Artery_Tibial 1.73E-5

rs1200345 TYRO3 rs1200345 1.000 Lung† 1.50E-24

rs2322659 AC093391.2 rs1030765 0.807 Artery_Tibial 3.82E-6

rs2322659 CCNT2 rs2322659 1.000 Lung† 2.37E-6

rs2322659 CXCR4 rs309134 0.837 Blood‡ 3.98E-6

rs2322659 DARS rs2322659 1.000 Blood‡ 1.06E-57

30

sentinel Gene eQTL SNP r2 with sentinel tissue P-value

rs2322659 1.000 Lung† 3.94E-26

rs2322659 1.000 Thyroid 3.48E-8

rs2304599 0.903 Artery_Tibial 2.46E-5

rs2322659

MCM6

rs2322659 1.000 Blood‡ 1.91E-119

rs2322659 1.000 Lung† 3.29E-10

rs2322659 1.000 Esophagus_Muscularis 3.40E-7

rs2322659 1.000 Thyroid 7.90E-7

rs2322659 1.000 Nerve_Tibial 1.46E-6

rs2322659 1.000 Artery_Aorta 2.55E-6

rs2322659

UBXN4

rs2322659 1.000 Skin_Sun_Exposed_Lower_leg 4.19E-8

rs1030764 0.908 Lung† 2.60E-6

rs2322659 ZRANB3 rs1470457 0.946 Lung† 4.36E-6

rs3849969

AGAP5

rs10740417 1.000 Testis 1.24E-6

rs10740417 1.000 Muscle_Skeletal 1.87E-5

rs3849969 CAMK2G rs3812637 1.000 Blood‡ 1.67E-5

rs3849969

FUT11

rs11000765 1.000 Nerve_Tibial 1.48E-5

rs10762556 0.995 Muscle_Skeletal 2.12E-5

rs10762556 0.995 Cells_Transformed_fibroblasts 3.86E-5

rs3849969 NDST2 rs3812637 1.000 Blood‡ 5.36E-47

rs3849969 PLAU,C10orf55 rs3812637 1.000 Blood‡ 1.57E-8

rs3849969 SEC24C rs3849969 1.000 Stomach 9.57E-7

rs6088813 CEP250 rs4911494 1.000 Blood‡ 4.14E-8

rs6088813

EDEM2

rs6088813 1.000 Cells_Transformed_fibroblasts 7.06E-6

rs1540927 0.996 Blood‡ 3.05E-5

rs6088813

EIF6

rs6141548 1.000 Blood‡ 3.96E-31

rs725908 0.996 Cells_Transformed_fibroblasts 3.12E-5

rs6088813 ERGIC3 rs4911496 1.000 Brain_Hippocampus 1.20E-6

rs6088813 FAM83C rs4911494 1.000 Skin_Sun_Exposed_Lower_leg 1.82E-6

rs6088813

GDF5

rs6142364 1.000 Lung 6.26E-9

rs4353719 1.000 Lung† 2.12E-7

rs6142360 0.996 Esophagus_Muscularis 2.95E-5

rs6088813 GGT7 rs6088817 1.000 Lung† 1.32E-5

rs6088813

RP3-477O4.16

rs6142349 1.000 Testis 1.16E-7

rs4911494 1.000 Skin_Sun_Exposed_Lower_leg 2.24E-6

rs6142364 1.000 Nerve_Tibial 4.63E-5

rs6088813

RPL36P4

rs6088813 1.000 Cells_Transformed_fibroblasts 2.39E-10

rs6088813 1.000 Adipose_Subcutaneous 6.89E-10

rs6088813 1.000 Nerve_Tibial 1.05E-6

rs1406948 0.937 Adrenal_Gland 1.20E-5

rs6088813 1.000 Thyroid 1.40E-5

rs6088813

UQCC1

rs6088809 1.000 Lung† 5.69E-24

rs6088813 1.000 Blood‡ 1.65E-11

rs6142364 1.000 Cells_Transformed_fibroblasts 1.59E-28

31

sentinel Gene eQTL SNP r2 with sentinel tissue P-value

rs6142364 1.000 Muscle_Skeletal 5.24E-11

rs6142364 1.000 Lung 1.69E-10

rs6088813 1.000 Nerve_Tibial 1.25E-8

rs6088813 1.000 Stomach 7.92E-6

rs6088813 1.000 Pancreas 1.31E-5

32

Supplementary Table 9: SIFT/Polyphen predictions for sentinel SNPs and proxies (r2>0.8). All sentinel SNPs and proxies (r2>0.8) were annotated using variant effect predictor and any SNP annotated as deleterious or damaging by either SIFT or PolyPhen identified.

Sentinel Proxy R2 Chr:Pos Allele Consequence Gene Transcript ID Codons SIFT consequence (score) PolyPhen consequence (score)

rs1200345 - - 15:41527518 T missense_variant RPAP1 ENST00000304330 Gag/Aag deleterious(0) possibly_damaging(0.526)

rs1200345 - - 15:41527518 T missense_variant, NMD_transcript_variant RPAP1 ENST00000562303 Gag/Aag deleterious(0.01) possibly_damaging(0.718)

rs1200345 - - 15:41527518 T missense_variant RPAP1 ENST00000304330 Gag/Aag deleterious(0) possibly_damaging(0.526)

rs1200345 - - 15:41527518 T missense_variant RPAP1 ENST00000561603 Gag/Aag tolerated(0.05) probably_damaging(0.996)

rs1200345 - - 15:41527518 T missense_variant, NMD_transcript_variant RPAP1 ENST00000562303 Gag/Aag deleterious(0.01) possibly_damaging(0.718)

rs6088813 rs224331 0.895 20:35434589 A missense_variant GDF5 ENST00000374369 Gcc/Tcc tolerated(0.8) possibly_damaging(0.455)

rs6088813 rs224331 0.895 20:35434589 A missense_variant GDF5 ENST00000374372 Gcc/Tcc tolerated(0.8) possibly_damaging(0.455)

rs6088813 rs224331 0.895 20:35434589 A missense_variant GDF5OS ENST00000374375 gCg/gAg tolerated_low_confidence(0.59) possibly_damaging(0.773)

rs6088813 rs224331 0.895 20:35434589 T missense_variant GDF5OS ENST00000374375 gCg/gTg tolerated_low_confidence(0.1) possibly_damaging(0.691)

33

Supplementary Table 10: Protein and RNA expression results all implicated genes from the single variant association analyses. Implicated genes were those located at, or close to the position of the sentinel SNP, or were implicated through eQTL data (See Supplementary Table 10). Protein expression are qualitative antibody based protein profiles in the respiratory system for epithelial cells, pneumocytes and macrophages from the Human Protein Atlas. RNA expression is quantitative data estimating the transcript abundance of each protein-coding gene by RNA-seq from the Human Protein Atlas and GTEx. FPKM: Fragmments Per Kilobase gene model and Million reads. RPKM: Reads Per Kilobase gene model and Million mapped reads.

Protein Expression RNA Profile

Human Protein Atlas Human Protein Atlas GTEx

Gene nasopharynx bronchus lung macrophages

lung pneumocytes

FKPM Category RPKM Category

LCT low medium medium not detected 0.0 not detected 0 not detected

FGF10 medium medium low low 3.0 low 1.4 low

LY86 not detected not detected high not detected 18.6 medium 7.4 low

SEC24C medium medium medium medium 24.4 medium 30.7 medium

RPAP1 high high medium high 4.4 low 4.5 low

CASC17 NA NA NA NA NA NA NA NA

UQCC1 medium high medium medium 11.5 medium 6.4 low

NDST2 not detected not detected not detected not detected 12.1 medium 0.8 low

PLAU low NA low low 15.1 medium 14 medium

C10orf55 not detected low not detected not detected 0.1 not detected 0 not detected

CAMK2G low medium not detected not detected 16.8 medium 10.2 medium

FUT11 medium medium medium high 7.1 low 5.5 low

AGAP5 medium medium not detected low 1.1 low 2.2 low

SPTBN5 NA NA NA NA 0.3 not detected 1 low

LTK medium medium medium low 9.5 low 13.6 medium

NDUFAF1 medium medium medium medium 10.4 medium 6 low

MCM6 medium medium not detected medium 5.4 low 7.2 low

DARS high medium medium medium 29.1 medium 17.8 medium

CXCR4 NA NA NA NA 23.1 medium 59.2 high

AC093391.2 NA NA NA NA NA NA NA NA

UBXN4 high high high medium 27.8 medium 29.5 medium

CEP250 NA NA NA NA 5.9 low 3.3 low

EIF6 medium NA medium low 45.5 medium 39.3 medium

EDEM2 medium medium medium medium 20.2 medium 11.9 medium

RPL36P4 NA NA NA NA NA NA NA NA

ERGIC3 medium medium medium low 87.5 high 52.2 high

GDF5 NA NA NA NA 0.7 low 1.1 low

RP3-477O4.16 NA NA NA NA NA NA NA NA

FAM83C NA NA NA NA NA NA NA NA

JMJD7-PLA2G4B low high medium not detected 3.8 low NA NA

MAPKBP1 medium medium low not detected 3.8 low 4.9 low

TYRO3 medium medium high not detected 2.4 low 4.1 low

CCNT2 high high high high 10.3 medium 8.9 low

ZRANB3 medium medium medium medium 0.6 low 0.3 not detected

34

GGT7 high medium medium not detected 7.8 low 10.9 medium

35

Supplementary Table 11: Look-up of association results for SNPs at 7 of the 12 loci which showed allele frequency differences between individuals from different regions in the UK. Shown are the P-values for associations with the three lung function traits (FEV1, FVC and FEV1/FVC), in both the analyses restricted to EA individuals, or

in the analysis of EA and AA individuals combined. The 7 loci shown are those with an available proxy (r2>0.2) for the SNPs reported in 54.

Discovery analysis P-values

Discovery Analysis SNP Population structure SNP EA + AA EA only

SNP CHR POS SNP r2 with analysed SNP

FEV1 FVC FEV1/FVC FEV1 FVC FEV1/FVC

rs3754689 2 136590746 rs1042712 0.9093 0.066 1.95E-03 0.216 0.011 3.19E-04 0.433

rs4331786 4 38769408 rs7696175 0.3641 0.250 0.168 0.457 0.254 0.169 0.415

rs9378805 6 417727 rs9378805 1 0.250 0.130 0.765 0.191 0.109 0.913

rs3873379 6 31262169 rs3873375 1 0.406 0.242 0.786 0.464 0.360 0.799

rs755383 9 863635 rs11790408 0.2032 0.960 0.538 0.169 0.764 0.414 0.196

rs12785878 11 71167449 rs12797951 0.9259 0.484 0.686 0.471 0.526 0.825 0.404

rs61976859 14 31583512 rs17449560 0.2756 0.753 0.337 0.433 0.680 0.343 0.530

36

Supplementary Table 12: All traits results for the seven novel lung function loci. Two-sided P-values are given for the discovery stage meta-analysis consisting of up to 68,470 individuals from the SpiroMeta and CHARGE Consortia.

SNP Chr:Pos (Nearest) Gene(s) Smoking Ancestry Effect / other allele

Effect Allele Frequency (Discovery)

FEV1 FVC FEV1/FVC

meta.Z meta.P meta.Z meta.P meta.Z meta.P

rs2322659 2:136555659 LCT (nonsynonymous) All Individuals

EA Only T/C 23.50% 4.5938 4.35E-06 5.5968 2.18E-08 -0.5893 0.556

rs1448044 5:44296986 FGF10 (dist=8111), NNT (dist=591,318)

Ever Smokers

EA+AA A/G 35.60% 2.9427 3.25E-03 4.8129 1.49E-06 -1.4533 0.146

rs1294421 6:6743149 LY86 (dist=87933),

RREB1 (dist=364,681) All

Individuals EA+AA T/G 36.80% -1.4521 0.146 1.6555 0.098 -5.4795 4.27E-08

rs3849969 10: 75525999 SEC24C (intronic) All

Individuals EA+AA T/C 29.40% 4.7665 1.87E-06 3.5369 4.05E-04 3.0351 2.40E-03

rs1200345 15: 41819716 RPAP1

(nonsynonymous) All

Individuals EA only C/T 48.80% -0.9685 0.333 1.3016 0.193 -4.5865 4.508E-06

rs1859962 17: 69108753 CASC17 (intronic) All Individuals

EA only G/T 48.20% 4.8760 1.08E-06 3.4002 6.73E-04 3.2432 1.18E-03

rs6088813 20: 33975181 UQCC1 (intronic) All Individuals

EA+AA C/A 36.70% -3.5529 3.81E-04 -4.6343 3.582E-06 1.3266 0.185

37

Supplementary Table 13: Genomic Inflation Factors: consortium and meta-analysis level. The SpiroMeta analysis included EA individuals only, thus the l values for EA only and EA+AA are identical.

Trait Smoking Ancestry SpiroMeta genomic inflation factor (l)

CHARGE l Meta-analysis l

FEV1 All Individuals EA only 1.060 1.178 1.038

FEV1 All Individuals EA+AA 1.060 1.169 1.048

FEV1 Ever Smokers EA only 1.023 1.054 1.028

FEV1 Ever Smokers EA+AA 1.023 1.061 1.027

FEV1 Never Smokers EA only 1.020 1.040 1.018

FEV1 Never Smokers EA+AA 1.020 1.053 1.033

FVC All Individuals EA only 1.072 1.146 1.037

FVC All Individuals EA+AA 1.072 1.142 1.039

FVC Ever Smokers EA only 1.001 1.066 1.008

FVC Ever Smokers EA+AA 1.001 1.070 1.014

FVC Never Smokers EA only 1.031 1.029 1.046

FVC Never Smokers EA+AA 1.031 1.028 1.040

FEV1/FVC All Individuals EA only 1.037 1.135 1.039

FEV1/FVC All Individuals EA+AA 1.037 1.136 1.042

FEV1/FVC Ever Smokers EA only 1.018 1.094 1.031

FEV1/FVC Ever Smokers EA+AA 1.018 1.104 1.034

FEV1/FVC Never Smokers EA only 1.006 1.036 1.022

FEV1/FVC Never Smokers EA+AA 1.006 1.016 1.017

38

References

1. The Atherosclerosis Risk in Communities (ARIC) Study: design and objectives. The ARIC investigators. Am J Epidemiol 1989; 129(4): 687-702.

2. Strachan DP, Rudnicka AR, Power C, et al. Lifecourse influences on health among British adults: effects of region of residence in childhood and adulthood. Int J Epidemiol 2007; 36(3): 522-31.

3. Marossy AE, Strachan DP, Rudnicka AR, Anderson HR. Childhood chest illness and the rate of decline of adult lung function between ages 35 and 45 years. Am J Respir Crit Care Med 2007; 175(4): 355-9.

4. Fried LP, Borhani NO, Enright P, et al. The cardiovascular health study: design and rationale. Ann

Epidemiol 1991; 1(3): 263-76.

5. Enright PL, Kronmal RA, Higgins M, Schenker M, Haponik EF. Spirometry reference values for women and men 65 to 85 years of age. Am Rev Respir Dis 1993; 147(125): e133.

6. Zemunik T, Boban M, Lauc G, et al. Genome-wide association study of biochemical traits in Korcula Island, Croatia. Croatian medical journal 2009; 50(1): 23-33.

7. Higgins M, Province M, Heiss G, et al. NHLBI Family Heart Study: objectives and design. Am J

Epidemiol 1996; 143(12): 1219-28.

8. Smith BH, Campbell A, Linksted P, et al. Cohort Profile: Generation Scotland: Scottish Family Health Study (GS:SFHS). The study, its participants and their potential for genetic research on health and illness. Int J Epidemiol 2013; 42(3): 689-700.

9. Waterer GW, Wan JY, Kritchevsky SB, et al. Airflow Limitation is Underrecognized in Well-Functioning Older People. J Am Geriatr Soc 2001; 49(8): 1032-8.

10. Thuesen BH, Cerqueira C, Aadahl M, et al. Cohort Profile: the Health2006 cohort, research centre for prevention and health. Int J Epidemiol 2014; 43(2): 568-75.

11. Jorgensen T, Borch-Johnsen K, Thomsen TF, Ibsen H, Glumer C, Pisinger C. A randomized non-pharmacological intervention study for prevention of ischaemic heart disease: baseline results Inter99. Eur J Cardiovasc Prev Rehabil 2003; 10(5): 377-86.

12. Taylor Jr HA, Wilson JG, Jones DW, et al. Toward resolution of cardiovascular health disparities in African Americans: design and methods of the Jackson Heart Study. Ethn Dis 2005; 15(4 Suppl 6): S6-4.

13. American Thoracic Society. Standardization of spirometry: 1994 update. Am J Respir Crit Care

Med 1995; 152: 1107-36.

39

14. Holle R, Happich M, Lowel H, Wichmann HE, Group MKS. KORA--a research platform for population based health research. Gesundheitswesen 2005; 67 Suppl 1: S19-25.

15. Wichmann HE, Gieger C, Illig T, Group MKS. KORA-gen--resource for population genetics, controls and a broad spectrum of disease phenotypes. Gesundheitswesen 2005; 67 Suppl 1: S26-30.

16. Miller MR, Hankinson J, Brusasco V, et al. Standardisation of spirometry. Eur Respir J 2005; 26(2): 319-38.

17. Peters A, Doring A, Ladwig KH, et al. [Multimorbidity and successful aging: the population-based KORA-Age study]. Zeitschrift fur Gerontologie und Geriatrie 2011; 44 Suppl 2: 41-54.

18. Deary IJ, Gow AJ, Taylor MD, et al. The Lothian Birth Cohort 1936: a study to examine influences on cognitive ageing from age 11 to age 70 and beyond. BMC Geriatr 2007; 7: 28.

19. Bild DE, Bluemke DA, Burke GL, et al. Multi-Ethnic Study of Atherosclerosis: objectives and design. Am J Epidemiol 2002; 156(9): 871-81.

20. Kaufman JD, Adar SD, Allen RW, et al. Prospective study of particulate air pollution exposures, subclinical atherosclerosis, and clinical cardiovascular disease: The Multi-Ethnic Study of Atherosclerosis and Air Pollution (MESA Air). Am J Epidemiol 2012; 176(9): 825-37.

21. Pellegrino R, Viegi G, Brusasco V, et al. Interpretative strategies for lung function tests. Eur

Respir J 2005; 26(5): 948-68.

22. Hankinson JL, Kawut SM, Shahar E, Smith LJ, Stukovsky KH, Barr RG. Performance of American Thoracic Society-recommended spirometry reference values in a multiethnic sample of adults: the multi-ethnic study of atherosclerosis (MESA) lung study. CHEST Journal 2010; 137(1): 138-45.

23. Rodriguez J, Jiang R, Johnson WC, MacKenzie BA, Smith LJ, Barr RG. The association of pipe and cigar use with cotinine levels, lung function, and airflow obstruction: a cross-sectional study. Ann

Intern Med 2010; 152(4): 201-10.

24. Lind L, Fors N, Hall J, Marttala K, Stenborg A. A comparison of three different methods to evaluate endothelium-dependent vasodilation in the elderly: the Prospective Investigation of the Vasculature in Uppsala Seniors (PIVUS) study. Arteriosclerosis, thrombosis, and vascular biology

2005; 25(11): 2368-75.

25. Hofman A, Brusselle GG, Murad SD, et al. The Rotterdam Study: 2016 objectives and design update. Eur J Epidemiol 2015; 30(8): 661-708.

26. Celli BR, MacNee W, ATS/ERS Task Force. Standards for the diagnosis and treatment of patients with COPD: a summary of the ATS/ERS position paper. Eur Respir J 2004; 23(6): 932-46.

40

27. Martin BW, Ackermann-Liebrich U, Leuenberger P, et al. SAPALDIA: methods and participation in the cross-sectional part of the Swiss Study on Air Pollution and Lung Diseases in Adults. Sozial- und

Praventivmedizin 1997; 42(2): 67-84.

28. Ackermann-Liebrich U, Kuna-Dibbert B, Probst-Hensch NM, et al. Follow-up of the Swiss Cohort Study on Air Pollution and Lung Diseases in Adults (SAPALDIA 2) 1991-2003: methods and characterization of participants. Sozial- und Praventivmedizin 2005; 50(4): 245-63.

29. John U, Greiner B, Hensel E, et al. Study of Health In Pomerania (SHIP): a health examination survey in an east German region: objectives and design. Soz Praventivmed 2001; 46(3): 186-94.

30. Volzke H, Alte D, Kohlmann T, Ludemann J, Nauck M, John U, Meng W. Reference intervals of serum thyroid function tests in a previously iodine-deficient area. Thyroid 2005; 15(3): 279-85.

31. Nelson SB, Gardner RM, Crapo RO, Jensen RL. Performance evaluation of contemporary spirometers. Chest 1990; 97(2): 288-97.

32. Standardization of spirometry--1987 update. Statement of the American Thoracic Society. Am

Rev Respir Dis 1987; 136(5): 1285-98.

33. Wain LV, Shrine N, Miller S, et al. Novel insights into the genetics of smoking behaviour, lung function, and chronic obstructive pulmonary disease (UK BiLEVE): a genetic association study in UK Biobank. Lancet Respir Med 2015; 3(10): 769-81.

34. Raitakari OT, Juonala M, Ronnemaa T, et al. Cohort profile: the cardiovascular risk in Young Finns Study. International journal of epidemiology 2008; 37(6): 1220-6.

35. Mahajan A, Robertson N, Rayner W. Exome-Chip Quality Control SOP. Version 5, 2012-11-20. 2012; .

36. Illumina Inc. Illumina GenCall Data Analysis Software. 2005; .

37. Goldstein JI, Crenshaw A, Carey J, et al. zCall: a rare variant caller for array-based genotyping: Genetics and population analysis. Bioinformatics 2012; 28(19): 2543-5.

38. Grove ML, Yu B, Cochran BJ, et al. Best practices and joint calling of the HumanExome BeadChip: the CHARGE Consortium. PLoS One 2013; 8(7): e68095.

39. Wain LV, Shrine N, Soler Artigas M, et al. Genome-wide association analyses for lung function and chronic obstructive pulmonary disease identify new loci and potential druggable targets. 2017; .

40. The 1000 Genomes Project Consortium. An integrated map of genetic variation from 1,092 human genomes. Nature 2012; 491(7422): 56-65.

41. Huang J, Howie B, McCarthy S, et al. Improved imputation of low-frequency and rare variants using the UK10K haplotype reference panel. Nat Commun 2015; 6: 8111.

41

42. Westra H, Peters MJ, Esko T, et al. Systematic identification of trans eQTLs as putative drivers of known disease associations. Nat Genet 2013; 45(10): 1238-43.

43. Lonsdale J, Thomas J, Salvatore M, et al. The Genotype-Tissue Expression (GTEx) project. Nat

Genet 2013; 45(6): 580-5.

44. Obeidat M, Miller S, Probert K, et al. GSTCD and INTS12 Regulation and Expression in the Human Lung. PLoS ONE 2013; 8(9): e74630.

45. Lamontagne M, Couture C, Postma DS, et al. Refining susceptibility loci of chronic obstructive pulmonary disease with lung eqtls. PLoS One 2013; 8(7): e70220.

46. Hao K, Bossé Y, Nickle DC, et al. Lung eQTLs to help reveal the molecular underpinnings of asthma. PLoS Genet 2012; 8(11): e1003029.

47. Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, Speed TP. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics

2003; 4(2): 249-64.

48. Soler Artigas M, Loth DW, Wain LV, et al. Genome-wide association and large-scale follow up identifies 16 new loci influencing lung function. Nat Genet 2011; 43(11): 1082-90.

49. Cho MH, Boutaoui N, Klanderman BJ, et al. Variants in FAM13A are associated with chronic obstructive pulmonary disease. Nat Genet 2010; 42(3): 200-2.

50. Repapi E, Sayers I, Wain LV, et al. Genome-wide association study identifies five loci associated with lung function. Nat Genet 2010; 42(1): 36-44.

51. Soler Artigas M, Wain LV, Miller S, et al. Sixteen new lung function signals identified through 1000 Genomes Project reference panel imputation. Nat Commun 2015; 6: 8658.

52. Loth DW, Soler Artigas M, Gharib SA, et al. Genome-wide association analysis identifies six new loci associated with forced vital capacity. Nat Genet 2014; 46(7): 669-77.

53. Pillai SG, Ge D, Zhu G, et al. A Genome-Wide Association Study in Chronic Obstructive Pulmonary Disease (COPD): Identification of Two Major Susceptibility Loci. PLoS Genet 2009; 5(3): e1000421.

54. Burton PR, Clayton DG, Cardon LR, et al. Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature 2007; 447(7145): 661-78.